Substituting Diacritics

[Gulliver]

In German, ö, ä, and ü can be replaced with oe, ae and ue and ß to ss when umlauts and ligatures aren't available (like when using a keyboard layout that doesn't support them).

In French, they are just left out (capital letters are often un-accented), if one is typing quickly and it is informal (facebook etc).

In English, use of ligatures has died down to almost non-existent, but many people don't know how to type them anyway. And beautiful Latin diacritic abbreviations are long gone...

Soooo... how do other languages handle accents when keyboards do not permit easy access?

[Ulrike Meinhof]

Swedish å, ä and ö are usually just written a, a and o when they're not available. Some people use aa, ae and oe, but that's much harder to read.

[FSI]

In Romanian they're usually left out completely in informal writing. However, you can usually spot some "tz"s and "sh"s to differentiate between words, e.g. " fată - faţă " girl - face ; " mâţă - mată - mata " cat - mat/opaque - you , and such like.

It can usually lead to ambiguous phrases and words and when I read such texts I find it sometimes hard to anticipate where the accent falls, which is rather important in verbs (so I know what tense it is). I think with just a few modifications Romanian could be very well written without any diacritics whatever.

[Magb]

Swedish å, ä and ö are usually just written a, a and o when they're not available. Some people use aa, ae and oe, but that's much harder to read.

That's interesting, since Norwegians use <aa ae oe> more often than not, in my experience.

[Jipí]

In German, ö, ä, and ü can be replaced with oe, ae and ue and ß to ss when umlauts and ligatures aren't available (like when using a keyboard layout that doesn't support them).

TBH, I sometimes find it harder to read if words are spelled with the umlauts resolved. Even if you leave out umlauts words are mostly clear from context. That umlauts turn into V+e and ß resolves to ss if not available has historic reasons: They were spelled aͤoͤuͤ (i.e. with superscript es, written like // in Kurrent cursives like Sütterlin) and, respectively, ſʒ/ſs in blackletter. However, please never replace ß with either B or β, that's just typographically wrong and awkward looking. Also, there is ẞ (capital es-zett) now, but traditionally, ß is turned into SS in all- and small caps. Since it's been introduced to Unicode I've never seen ẞ in use, actually. GROẞ and GROSS both look alright. I guess it's only actually useful to distinguish the spelling of names. The difference between Masse and Maße and Busse and Buße should be clear from context mostly.

Fun fact: In Basel, Switzerland, there's an Aeussere Baselstrasse (although they use ÄÖÜ in Switzerland). POW! In FRGian spelling that's Äußere Baselstraße.

[Bedelato]

Can we talk about conlangs here?

[Aurora Rossa]

Can we talk about conlangs here?

If so, I wouldn't mind chiming in. Diacritics in the Terpish script form an integral part of the characters to which they join, pretty much like the vowel diacritics in South Asian scripts. For that reason you can't really omit them anymore than you could omit entire pieces of letters when writing with the Roman script. Of course, Terpish keyboards were designed specifically for the Terpish script so they never have problems with lacking appropriate keys for diacritics.

[Jipí]

Well, re: my conlang, I think you can just double vowels with a macron. I don't use any other diacritics besides that. In my con-script, though, not using any diacritics would not work.

[Bedelato]

The reason I asked was because Bengedian does something like this. In romanized Bengeian the letters <š č> /ʃ t͡ʃ/ are written <sh ch> when the keyboard layout or character encoding doesn't support them. Accents would just be left out if they were unsupported.

My conscript has seperate letters for those phonemes though, so this problem would be avoided had computers been designed by Bengedian speakers instead of English speakers.

[Izambri]

Catalan uses accents and diaeresis. When they're not available vowels are written without them.

[Qwynegold]

I've seen both <a e> and <ae oe> been used in place of <ä ö> in Finnish.

I just took the Universal Declaration of Human Rights, divided it into two parts and replaced <ä ö> with <a o> in one, and with <ae oe> in the other. They were both easy to read. I had a few difficulties with <a o>, mostly in words I'm not used to. <ae oe> didn't pose any such difficulties, but it was somewhat cumbersome to read, especially with words like "paeaemaeaeriae".

[tezcatlip0ca]

"paeaemaeaeriae"

Try "paaemaaeriae". After all, /ɑæ/ sequences do not occur in Finnish.

[tezcatlip0ca]

And finnish this once and for all!

[alice]

"paeaemaeaeriae"

Try "paaemaaeriae". After all, /ɑæ/ sequences do not occur in Finnish.

What's wrong with a bit of redundancy? How about "paaemaaria", where you only indicate frontness on the first affected vowel? This might at least prolong the lives of all those umlaut keys on Finnish keyboards!

[Qwynegold]

I just used search-and-replace.

[Åge Kruger]

Swedish å, ä and ö are usually just written a, a and o when they're not available. Some people use aa, ae and oe, but that's much harder to read.

That's interesting, since Norwegians use <aa ae oe> more often than not, in my experience.

I'm not so sure. It may well vary from medium to medium, but domain names tend to replace ø with o, æ with ae, and å with a (actually, a/aa is pretty even, I reckon, but I think most people would try with just a first).

[Magb]

Swedish å, ä and ö are usually just written a, a and o when they're not available. Some people use aa, ae and oe, but that's much harder to read.

That's interesting, since Norwegians use <aa ae oe> more often than not, in my experience.

I'm not so sure. It may well vary from medium to medium, but domain names tend to replace ø with o, æ with ae, and å with a (actually, a/aa is pretty even, I reckon, but I think most people would try with just a first).

That is indeed true of domain names, and maybe proper nouns in general. I was mostly thinking of emails, comments/posts and such. My "more often than not" statement could very well be inaccurate.

[Nortaneous]

Estonian replaces o-tilde and o-umlaut with <6> and <8>. Not sure what they do with the other letters with diacritics.

[Apeiron]

One option is inline diacritics:

o:
n~
,c
e'
d^

Before for below, after for above.

[gsandi]

Hungarians tend to leave of diacritics in e-mails, which can make reading their e-mails idfficult, but rarely incomprehensible. Context helps enormously.

They may leave off diacritics even if most are available (the ones that are hardest to locate are the long umlauts: Ő ő Ű ű), as there is a problem with high-bit ASCII codes being translated differently in different e-mail programs.

[Shrdlu]

My conlang only use an diacrit to distinguse between the close central rounded vowel /u/ and the close back rounded vowel,./ú/. That's all.

[Lyhoko Leaci]

Zukish replaces ä with ei, ë with ie, ï with ai, ö with ou, ü with yu, þ with tc, and ð with dc.

Myonian replaces þ with th, ð with dh, and ŋ with ng.

However, as the keyboards used there were developed independently of the ones on Earth, they already have the necessary keys.

[Declan]

Estonian replaces o-tilde and o-umlaut with <6> and <8>. Not sure what they do with the other letters with diacritics.

Is that not really hard to read? Or is it something that's relatively easy to get used to?

[Acid Badger]

Estonian replaces o-tilde and o-umlaut with <6> and <8>. Not sure what they do with the other letters with diacritics.

Is that not really hard to read? Or is it something that's relatively easy to get used to?

I don't know about Estonian, but I know how some greek friends of mine write Greek in text messages, replacing ξ with 3, θ with 8, η with h, χ with x and so on. Looks a little bit like leetspeak, but they are able to read it.

[Bristel]

Squalipsh is replete with diacritic marks.

Code: Select all

stop: /p p’ t t’ k k’ kʷ k’ʷ q qʷ q’ q’ʷ ʔ/ <p p’ t t’ k k’ kʷ kʷ’ q qʷ q’ qʷ’ ʔ>
affricate: /t͡s’ t͡ɬ’ t͡ʃ’/ <ts’ tł’ tš’>
fricative: /s ʃ ɬ xʷ χ χʷ h/ <s š ł xʷ ẋ ẋʷ h>
nasal: /m mˀ n nˀ/ <m ṁ n ṅ>
approx.: /l lˀ j jˀ w wˀ/  <l l’ y y’ w ẇ>
vowels: /a e i o u ə/ <a e i o u ă>

I haven't thought of a diacritic-free version of the orthography yet...

But I think it would be full of ambiguities.

Here's one I just came up with, minus the plain consonants and vowels:

<p' t' k' kw' qw q' qw' '>
<ts' tlh' tsh'>
<sh lh xw xh xhw>
<l' m' n'>
<a>

There's definitely a chance of ambiguity, especially with the glottal stop or /a/ vs. /ə/.